High-Quality Epitaxial Growth of Mg₃Sb₂ Thin Films: Enhancing Thermoelectric Performance through Optimized Substrate and Growth Conditions

Thermoelectric materials have a remarkable role to play in a cleaner energy future, offering a way to capture and convert wasted heat—think of the excess from factories or vehicle exhausts—directly into electricity. But to be genuinely practical for everyday use, these materials need to work effectively at room temperature, particularly in compact, eco-friendly designs. One of the critical benchmarks for assessing how well a thermoelectric material performs is its figure of merit, called ZT. Essentially, ZT reflects how well a material balances three key properties: a high Seebeck coefficient, solid electrical conductivity, and minimal thermal conductivity. Among the materials scientists are exploring, magnesium antimonide (Mg₃Sb₂) has emerged as a strong contender thanks to its excellent thermoelectric properties, cost-effectiveness, and a much lower toxicity profile compared to older materials like bismuth telluride (Bi₂Te₃). However, the road to achieving high-performance Mg₃Sb₂ in thin-film form isn’t without its challenges. Thin films are preferred for applications in tiny devices and compact thermoelectric modules, but their effectiveness can vary greatly depending on how well-aligned their crystal structure is and how free they are of imperfections like polycrystalline boundaries or random defects. These flaws can interrupt the flow of electrons, lowering electrical conductivity and reducing overall efficiency. Mg₃Sb₂ thin films, in particular, have been known to develop a polycrystalline or passive layer when directly grown on a substrate, which introduces scattering sites that limit electron movement and reduce the material’s thermoelectric potential. To tackle these issues head-on, a team of researchers recently published a study in Applied Physics Express. Akito Ayukawa, Nozomu Kiridoshi, Professor Haruhiko Udono, and led by Professor Shunya Sakane from Ibaraki University, along with Wakaba Yamamoto and Akira Yasuhara from JEOL Ltd. in Tokyo, the team developed innovative methods for growing high-quality Mg₃Sb₂ thin films with minimal defects and a highly consistent, epitaxial alignment. They focused on the use of annealed c-plane aluminum oxide (Al₂O₃) substrates, which could play a crucial role in refining both the uniformity and crystallographic orientation of the Mg₃Sb₂ films. By annealing the Al₂O₃ substrate, they created a smoother surface with neatly defined atomic steps, which in turn helps align the Mg₃Sb₂ lattice more effectively. This process minimizes the defects that could otherwise disrupt electron flow.

The research team started with an essential but often overlooked detail: the preparation of the Al₂O₃ substrate. Instead of diving straight into layering, they first annealed these substrates—essentially heating them up to 1000°C in air. This wasn’t just a routine step; it was key to creating a super-smooth surface with tiny, well-formed steps at the atomic level. These steps help make sure that the thin film would form in a clean, orderly way, minimizing the risk of defects that could throw off the alignment of the Mg₃Sb₂ crystals. Once the substrates were set up, the authors moved to the actual process of growing the Mg₃Sb₂ film using molecular beam epitaxy, or MBE. Now, MBE isn’t just any growth method—it’s known for its precision, allowing researchers to build up materials layer by layer with incredible control. They maintained an exact magnesium-to-antimony ratio of 4:1 and worked in an ultra-high vacuum to keep the film as pure as possible. This approach allowed the Mg₃Sb₂ layers to settle in a smooth, uniform structure directly on the Al₂O₃, avoiding any of the mixed or passive layers that usually disrupt electron flow and reduce efficiency. But growing the film was only half the battle. They needed to make sure it was actually aligned and defect-free. To keep tabs on the structure as it formed, they used reflection high-energy electron diffraction to track changes in the film’s pattern in real time. They watched as the initial patterns from the Al₂O₃ transitioned smoothly into patterns unique to the Mg₃Sb₂, indicating that the growth was going exactly as planned. For a closer look, they used atomic force microscopy, which showed a smoothness level with an impressively low roughness—root mean square roughness~0.44 nm. Further imaging with high-resolution transmission electron microscopy gave them a crisp view of the interface between the Al₂O₃ and the Mg₃Sb₂, showing a direct, clean attachment without any unwanted polycrystalline layers—a result that marked a major win for achieving high electron mobility and thermoelectric efficiency.

Next, the authors tested the thermoelectric performance of their Mg₃Sb₂ film by measuring its Seebeck coefficient and electrical conductivity, crucial indicators of thermoelectric potential. Compared to previous studies, their film showed a substantial leap in power factor, nearly three times higher than what others had found. This improvement was largely due to the high carrier mobility they achieved, with an average of 38 cm² V⁻¹ s⁻¹—far better than any prior results. The smooth, defect-free interface played a huge role in this, reducing any scattering that might hinder the flow of electrons. To confirm that the alignment stayed consistent throughout the film, they used X-ray diffraction, which verified the film’s strict c-plane orientation. This orientation is especially valuable for thermoelectric materials since it helps minimize heat loss and maximize electrical conductivity. For a more detailed look into the film’s properties, the authors compared the Seebeck coefficient with the carrier concentration, applying the single parabolic band model. They found that the high Seebeck coefficient aligned well with theoretical expectations for Mg₃Sb₂’s Zintl phase structure. Their work opens up promising directions for developing better thermoelectric materials and highlights how even small adjustments in film growth can lead to big gains in efficiency.

We believe the importance of the study by Professor Shunya Sakane and his colleagues lies in how it demonstrates, quite effectively, that high-quality, defect-free Mg₃Sb₂ thin films can be grown with a precise c-plane orientation—leading to a marked improvement in thermoelectric performance. By using an annealed Al₂O₃ substrate and fine-tuning growth conditions with molecular beam epitaxy, the researchers achieved something impressive: they addressed the long-standing challenges that have limited Mg₃Sb₂’s efficiency in thin-film form. The research work of Professor Shunya Sakane and colleagues shows a practical way to boost carrier mobility and reduce scattering by removing passive layers and polycrystalline boundaries, which often block or disrupt the flow of electrons. Altogether, this research paves the way for the development of compact, efficient thermoelectric modules that could potentially power low-energy devices or capture wasted heat, especially at room temperature. And with Mg₃Sb₂ being more affordable than traditional materials, its performance here suggests it could become a key material for scalable, eco-friendly thermoelectric applications. But the impact goes beyond just Mg₃Sb₂. This approach to substrate prep and thin-film growth could be applied to other thermoelectric materials, especially those in the Zintl phase family. This study sets a strong example for improving similar compounds, emphasizing how vital crystallographic alignment and high-quality interfaces are for enhancing thermoelectric properties. The fact that this new method achieves high efficiency without relying on rare or expensive materials is also a game-changer, potentially driving forward more affordable thermoelectric devices for a variety of uses—from power generation in portable electronics to sustainable energy systems in industry.